PURPOSE: To evaluate ovarian cancer staging and tumor resectability wi
th computed tomography (CT) or magnetic resonance (MR) imaging. MATERI
ALS AND METHODS: Eighty-two women underwent CT (n = 43) or MR imaging
(n = 50); eleven of these 82 underwent both. Imaging was performed wit
hin 4 weeks of surgical staging. Radiologic, surgical, and histopathol
ogic findings were compared. RESULTS: Overall staging accuracy was sim
ilar for CT and MR imaging (77% [33 of 43] vs 78% [39 of 50]). Evaluat
ion of pelvic cancer extent was better with MR imaging than with CT. T
here was no difference in detection of abdominal disease. Most mesente
ric and small-bowel implants were not detected with either CT or MR im
aging. For CT, the positive predictive value for cancer nonresectabili
ty was 100% (three of three patients); the negative predictive value w
as 92% (37 of 40 patients). The positive and negative predictive value
s for MR imaging were 91% (10 of 11 patients) and 97% (38 of 39 patien
ts). CONCLUSION: While the staging accuracy of both CT and MR imaging
is only moderate, prediction of tumor resectability is excellent.